Available online at www.sciencedirect.

com

ScienceDirect
Procedia Engineering 169 (2016) 191 – 198

4th International Conference on Countermeasures to Urban Heat Island (UHI) 2016

Experimental Study on the Influence of Urban Water Body on

Thermal Environment at Outdoor Scale Model
Nedyomukti Imam Syafiia*, Masayuki Ichinosea, Nyuk Hien Wongb, Eiko Kumakuraa,
Steve Kardinal Jusufc and Kohei Chigusaa
a
Department of Architecture and Building Engineering, Tokyo Metropolitan University, Japan
b
Department of Building, National Univeresity of Singapore, Singapore
c
Sustainable Infrastructure Engineering (Building Service), Singapore Institute of Technology, Singapore

Abstract

Urban Heat Island (UHI) has become a common urban problem that may lead to human health issues and increase in air-
conditioning energy demand due to increase of air temperature. In order to reduce the UHI effect and improve the outdoor thermal
environment is necessary to assess the effect of various measures. Water bodies’ ability to regulate the microclimate arguably has
the potential to mitigate the effect. Water bodies commonly found in urban area are usually described as a permanent or temporary
collection of water in form of small stationary water or pond. These water bodies contribute to altering the surrounding thermal
environment due to its cooling effect, either by evaporation or transfer of heat between air and the water. In real urban condition,
however, the heterogeneity makes it difficult to assess the cooling benefits and to isolate the effects of individual parameter (such
as shape, surface area, wind condition or solar radiation) under the complex physic process involved in the urban meteorology. As
an alternative method, an outdoor measurement on a physically reduced scale model was conducted in summer. The outdoor-scaled
model consists of an array of 1.5 m concrete cubes and able to recreate thermal pattern similar to the actual urban condition. A
typical of commonly found water bodies in urban areas, pond, was installed in the scale model to have a better understanding on
potential benefits of water bodies in urban area. In this paper, the author present initial findings on the characteristic of surrounding
microclimate and its influential factors, particularly ambient air temperature near urban water body derived from the proposed
approach. The experiment result shows interesting patterns, which generally shows cooling effect during the day and even though
is limited it also happen during the night. Moreover, the present of more solar radiation and low wind conditions are found further
enhance the cooling effect.
© 2016 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
© 2016 The Authors. Published by Elsevier Ltd.
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer-review under responsibility of the organizing committee of the 4th IC2UHI2016

* Corresponding author. Tel.: +818095370681.

E-mail address: nedyomukti@ugm.ac.id

1877-7058 © 2016 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer-review under responsibility of the organizing committee of the 4th IC2UHI2016
doi:10.1016/j.proeng.2016.10.023
192 Nedyomukti Imam Syafii et al. / Procedia Engineering 169 (2016) 191 – 198

Keywords: Urban water bodies; cooling effect; air temperature; scale model

1. Introduction

Water bodies have the ability to adjust the surrounding microclimate. The temperature mitigating capacity of water
bodies in urban environment can potentially reduce energy consumption, increase outdoor thermal comfort and
mitigate the Urban Heat Island (UHI) effect. Air temperature near bodies of water is found different from that over
land owe to differences in the way water heat and cool. Water bodies are known as the best absorbers of radiation,
nevertheless show very little thermal response [1]. Owe to its transparency, large thermal capacity and volume, the
incident solar radiation is able to transmitted to considerable depths and be spread throughout large volume. Together
with unlimited water for evaporation, bodies of water create an efficient heat sink and further cool the surface layer.
Furthermore, based on the energy budget point of view, more evaporation increases latent heat (Qh) and affects the
energy partitioning of sensible heat (Qe) and the stored energy (Qs), in which reduction of Qe and decrease magnitude
of Qs, make its immediate surrounding air temperature lower.
Study on the temperature reduction due to water bodies have been conducted by many researchers utilizing various
methods. By remote sensing, researchers estimate the cooling effect by analyzing the surface temperature of different
land use in urban area [2,3,4] and shows cooling effect of up to 5.63°C if urban wetlands are compared to urban area
[3]. By collecting data from on-site measurement, Katayama found that the air temperature above the river was found
lower than the nearby street when the sea breeze was blowing [5]. Yet, similar study shows a difference about 3-5°C
in air temperature between the river and the city area on fine days in warmer seasons [6]. In the United Kingdom,
England, another on site measurement study [7] on the microclimate around a small urban river show significant
cooling over the river, with an average cooling of nearly 1°C during ambient conditions greater than 20°C which most
likely happen during the hottest time of the day. Above a water pond, furthermore, the air temperature is found 1°C
to 2°C lower than the average air temperature of the surrounding park during daytime and much lower with the
introduction of water spray [8]. Likewise, field measurement study in Singapore found that water features, such as
pond and water wall are able to reduce the air temperature up to 1.8°C during sunny clear day [9]. Using WRF
(Weather Research and Forecasting) meso-scale model, Theeuwes found that bigger lake tends to have a higher
influence within the city with the respect to wind direction, while smaller lake equally distributed, although have a
smaller cooling effect, may influence a much higher percentage of the city [10]. A recent studies utilizing numerical
simulation study shows that beside the ability to reduce the air temperature the presence of water bodies able reduce
the energy consumption [11] and with additional vegetation, it will also provide better outdoor thermal comfort [12].
All of these researches show a clear evidence of water bodies modify the thermal environment by cooling down the
surrounding air temperature through evaporation cooling and convection. However, most of these studies provided
less information on the factors influencing the effects. Other than that, the effects of these factors are not fully
understood yet in actual condition. Further research is needed to evaluate the effect of water bodies to its immediate
surrounding within urban setting while finding the distinguish factors which may influence the effect. In the real case,
finding an ideal condition for comparison purposes is difficult. Thus, this paper looks at the issues in more detail from
studying the effect of water bodies on its surrounding thermal environment by utilizing a scale model, namely
Comprehensive Outdoor Scale Model (COSMO).
The scale model was built with an objective to simulate an idealized urban thermal environment [13]. As it is
located outdoor and it is simpler than the real urban condition, such a scale model has the advantages of observing
various physical phenomena under real climate conditions and arranging comprehensive measurement on a relatively
uniform area, thus, the result is easier to interpret. Yet, the scale model ability to recreate thermal pattern similar to
the actual urban condition, shows the benefits of finding the impact of urban modification, especially water bodies,
on street level thermal environment [14,15,16,17]. To clarify an evaluate the effects of water bodies on the thermal
environment and the influence factor affecting the cooling effect of an urban canyon in typical summer days, an
experimental study with a simple pond design, were installed inside the COSMO.
 Nedyomukti Imam Syafii et al. / Procedia Engineering 169 (2016) 191 – 198 193

2. Experiment Details

2.1. Comprehensive Outdoor Scale Model (COSMO)

COSMO, a 1:5 scaled down relative to typical residential district in Japan, is a regular cubical array of 512 concrete
blocks located in Saitama Prefecture, Japan. During summer, the climate near the ground was characterized with
relative high air temperature; dryer air and South East prevailing wind condition. Each of the blocks are 1.5 m x 1.5
m x 1.5 m in sized, made from concrete, hollow inside and painted with dark grey paint. The blocks are arranged in
lattice-type with 1.5 m distances apart; resemble a low-rise urban setting.

2.2. Water pond and points of measurement

A 1.5 m wide, 6 m long and 10 cm deep water pond was constructed of bricks, waterproof sheet and painted with
dark grey waterproof paint in such it surface will have similar condition to the surface of the COSMO. Additionally,
seven points of measurements were installed to measure the temperature profile and fluctuations along a line parallel
to the long axis of the COSMO. The points were arranged relative to the downwind side of the South East (SE)
prevailing wind. Following this prevailing wind pattern, point 1 and 3 are placed in the edge of the pond, aiming to
capture the thermal condition “just before” and “just after” of the pond. Point 2 is placed in the middle of the pond,
aiming to see the average or general condition of the thermal environment of the water pond. Other four points are
placed beyond the pond, with 1.5 m distances apart to each other. Each point consists of type-T bare thermocouples
with a junction wire diameter of 0.01 inch (approx. 0.025 mm) and records with a sampling frequency of 1 second.
The thermocouple wires placed at 30 cm height from the ground, which represent the scaled down “average human
height”. To measure the background climate, one weather station was placed 2m above the COSMO canopy and one
pyranometer was placed just above the COSMO blocks. A reference case was developed nearby as the control. The
reference case, consist of 3 point of measurement, aims to see the difference between each point location within the
canopy, without any modification and act as a basis of comparison. All detailed plan and photos of the points
arrangement are shown in Fig. 1 and detail of measurements equipment are shown on Table 1.

a b

Fig. 1.(a) footage and (b) plan view of the experiment study at COSMO (note: UCL: Urban Canopy layer, RL: Roughness layer, ISL: Inertial
Sublayer)

Table 1. Measurement tools detail

Measured parameter
Air temperature
Global solar radiation
Background climate
Equipment detail Positions
Thermocouple (copper – constant, 0.025 mm d) Refer to Fig.2, height: 30 cm
Pyranometer (Eko MS-610) Refer to Fig.2, above the
COSMO blocks
Vaisala WXT520 Refer to Fig.2, height: 2 m
above the COSMO blocks
194 Nedyomukti Imam Syafii et al. / Procedia Engineering 169 (2016) 191 – 198

2.3. Time period and selected day for analysis

The experimental study was conducted during the summer period for almost 2 months from end of July to early
September in 2015. Due to the limitation of the air temperature sensor (durability issues), for the following analysis,
only hot days and relatively less data error was used. The selected day data and the analysis purposes together with
the interested points are shown on Table 2 below.

Table 2. Selected days and points of interest.

Analysis Selected days Points of interest
Spatial distribution 5 Aug All corresponding points
Influence of shadow and wind 31 Jul; 1,2,4 & 5 Aug Point no.2 (above the pond)

Due to the uniqueness of each hot day in terms of wind and solar condition and to have a clear and simple way to
show the air temperature trends and behavior changes happening on the experimental study, only one typical hot day
was presented on this paper. Other typical hot day, however, have very similar trends although does not really have
the exact same value.

3. Findings and Discussion

General analysis was carried out on each daily set of selected day data to evaluate trends and points of interest. For
a general overview, most of the data was presented in hourly average. Meanwhile, for more detailed views, the data
is 1-min averaged and visualized with moving average trend-lines.

3.1. Spatial distribution of air temperature

The graph in Fig. 2 shows the air temperature difference (ǻTr) for two points in the reference case, with Point 3
acts as the basis of comparison. The data was presented in hourly averaged with positive value, which means that the
air temperature at Point 3 is cooler than the other points. The graph suggests that COSMO canopy doesn't have very
similar background air temperature condition. Generally, the points near the edge of COSMO (Point 3) are found
cooler, compared to points toward the middle of COSMO (Point 1). However, Point 2 which is located in the
intersection of the canyon, the air temperatures are found to be the coolest, particularly during daytime. This might be
due to the abundant wind from multiple directions at the intersection as compared to the other points, which are
positioned between blocks and mainly exposed to only one prevailing wind direction. Although there is no statistically
analysis to weight the background dissimilarity in this paper, this condition will be considered for all future analysis.

Fig. 2. Air temperature difference (ǻTr) profile of air temperature points at the reference respectively from air temperature of points at reference
no.3 on August 5

Fig. 3 shows the temperature profile of one typical hot day (5th of August 2015). The collected data was hourly
averaged and visualized with daily minimum (min), daily average (ave) and daily maximum (max) trend-lines with
two different time set, day and night. Daytime in this region, as it characterizes with the present of sun, is from 6am
 Nedyomukti Imam Syafii et al. / Procedia Engineering 169 (2016) 191 – 198 195

to 6pm; and night time, is from 6pm to 6am, during this summer season. The temperature profile shows that during
the daytime, the mitigating effect of the pond is only clearly shown on the max value. It suggests that when the air
temperature was at its highest, the cooling effect was the most prominent. The max trend lines also show a reduction
of cooling impact on every span away from the pond. The temperature difference from the coolest point (Point 2 - just
above the water pond) with the highest temperature observed point (Point 6 - 4.5cm away from the edge of the pond)
could reach up to 2.6°C. Although it was relatively minor and limited, the reduction of air temperature can still be
seen until Point 5, which is 3 m away from the water pond. On the min trend lines, however, there is no obvious sign
of air temperature reduction. If we considered the background dissimilarity, the reduction of air temperature will be
more likely to be higher.
During the nighttime, although max trend lines show a similar pattern to the daytime graph, the average and min
trend lines shows an opposite pattern. The points near the pond seem to have higher air temperature, and declining
towards down wind. This finding might shows that, during the night time, the heat release from the water pond has a
slower rate as compared to the concrete surface.

Fig. 3. Daily minimum, average and maximum air temperature profile

For further analysis, each data values from the six points near the pond were compared to the values from point
no.1 at the reference case. If the values from the reference case represent the condition of no evaporation, the average
difference in air temperature (ǻTa) from this calculation could refer to the water bodies cooling effect. For more
detailed analysis, the data was averaged to 1min and visualized with moving average trend lines.
Overall, the graph on Fig. 4 shows that the ǻTa values are generally negative for all points. These may suggest
that the mitigating effect of urban water bodies might happen not only during the day but also during the night, even
though is limited. The graph also shows three distinguish patterns, which might show the influence of other parameters
to ǻTa. First pattern occurred during the day, which started with the sudden temperature changes in the morning due
to the availability of solar radiation and early low wind speed. With different rate of warming, all of the points seem
to experience an increase in air temperature. Later in the morning, solar radiation induced the evaporation process
lowered the air temperature. The reduction of air temperature continued throughout the day. The ǻTa, however, shows
random pattern during the first part of the day (6am - 12pm). Not until the solar radiation is at the highest, the wind
speed is more than 1m/s and the wind direction is relatively stable (flowing from SE), the ǻTa at points near the pond
(Point 1-3) shows lower values as compare to the points away from ponds. Generally, Point 2 (just above the pond)
seems to have the lowest value for the whole day, with ǻTa difference up to 1.4°C. The next observable pattern is
showed during the night. Generally, when solar radiation is absent and there is still wind moving from the upwind of
the pond (SE), almost all of the points near the water pond are still cooler compared to the reference. This pattern is
relatively similar to the daytime pattern; with the points near to the water pond have lower air temperatures,
nevertheless there was less fluctuation of air temperature. However, when the wind speed was low (less than 1 m/s),
the last pattern appears. During this period of time, the air temperature profile shows a much clear diminishing profile
as a reversed pattern. Point no.7 is found to be the coolest and Point 1-3 is the warmest, suggesting that during the
night, the pond cools more slowly as compared to the concrete surface, in agreement with the earlier finding.
196 Nedyomukti Imam Syafii et al. / Procedia Engineering 169 (2016) 191 – 198

Fig. 4. Daily air temperature difference (ǻTa) profile of air temperature of points at the pond case respectively from air temperature of point no.1
at reference case

3.2. Influence of surrounding building shade and wind condition

In the nature, bodies of water are known to be the best absorbers of thermal radiation and exhibit very little thermal
response [1], in which make the water bodies have less temperature fluctuation. This unique thermal behavior may
help to understand more on the influence of the shadow from surrounding building blocks and the wind condition. For
this purposes the selected daily air temperature data was compared to water temperature and evaluated with reference
to shadow and wind condition.

Fig. 5. Typical shadow and wind condition inside the COSMO

 Nedyomukti Imam Syafii et al. / Procedia Engineering 169 (2016) 191 – 198 197

Fig. 6 shows the temperature difference of air temperature and water temperature (ǻTw) in addition to global solar
radiation and wind condition. The data shown on Fig. 8 is hourly averaged data with positive value which means the
water temperatures are warmer than the air temperature and vice versa. Generally, the water temperatures are found
warmer during the night and cooler during the day. Due to exceptionally large thermal capacity (water specific heat
capacity is 4.185 kj/kg.K, while air specific heat capacity is 1.006 kj/kg.k) and transparency, water bodies allows more
short wave radiation being absorbs during the day and release it during the night, thus explained such pattern.
However, there is unusual and rather consistent pattern happening during the day, which seems related to the shade
of the surrounding building blocks. During the day, the COSMO canyon experience less shadow during a certain time,
in which these times also show the time where the canyon experiences the most direct solar radiation (refer to Fig. 5,
grey background). At this particular time, the difference of air temperature and water temperature changes its pattern
quite dramatically, creating an interesting trend. The ǻTw patterns show a sudden increase and then sudden decrease
after this period of time. If water bodies tend to have small temperature fluctuation, the sudden increase and decrease
pattern of ǻTw may show the other influencing factor affecting the intensity of evaporative cooling, which in this case
the present of building shade. Less shadow means that there is more direct solar radiation incident on the surface of
the water, hence enhancing the evaporation process. Further analysis shows that the intensity of this anomaly seems
vary and depend on the wind speed. During a relatively low wind speed (less than 1.5 m/s), the ǻTw could go up to
positive value, as shown on Aug 1, 2, & 4. With a simple mathematics calculation, the rate of air temperature increase
was found 0.63°C/hour in average. However, when the wind speed is more than 1.5 m/s, the ǻTw increase rate was
lower (0.35˚C/hour in average) and was likely to be less than zero. It suggests that the intensity of the anomaly, which
eventually pictures the intensity of the cooling effect, is more prominent on a relatively low wind speed condition.

Fig. 6. Temperature differences of air temperature and water temperature (ǻTw) with addition of global solar radiation and wind condition

4. Limitation and Future Possible Research

The use of regular array blocks may be not really representing the real urban condition and merely as a
simplification. However, its simplicity is allowing researcher to identify different parameters. Although the immediate
cooling benefits of one simple water pond was evaluated in this paper, further research can be done by varying the
sizes of the water area or simulating different features, such as water wall, water fountain or trees.
198 Nedyomukti Imam Syafii et al. / Procedia Engineering 169 (2016) 191 – 198

5. Conclusion

This initial research shows the thermal effect of water bodies and its potential to improve the urban microclimate
as well as to mitigate the UHI effect. This effect extends beyond the water bodies area, particularly affecting the
downwind area. The max temperature difference is found up to 2.6°C lower during the hottest time of the day and the
min temperature is found 0.7°C warmer during the coolest time of the day. Additionally, even within the dissimilarity
background (the edge of COSMO are cooler), the reduction of air temperature is still visible. If this effect is included
to the findings, the reduction of air temperature may show higher value. On daily average temperature, however, there
are no significant differences of air temperature relative to the distances. Another key finding is, generally, with the
right wind speed and sufficient solar radiation, water body in an urban canyon is able to effectively reduce the ambient
air temperature. Therefore, utilizing the water bodies in urban context can be one of practical way of improving the
thermal environment and benefits the urban area.

Acknowledgement

The authors are gratefully acknowledging the support of Prof. K. Narita and Prof. I. Misaka from the Nippon
Institute of Technology for the valuable discussions and the privilege to use COSMO, which without it, the present of
the experimental study could not have been completed.

Reference

[1] T.R Oke, Boundary Layer Climates, Routledge, London, 1987.

[2] X. L. Chen, H. M. Zhao, P. X. Li and Z. Y. Yin, Remote sensing image-based analysis of the relationship between urban heat island and land
use/cover changes, Remote Sensing of Environment 104 (2006) 133–146.
[3] P. Hou, W. Jiang, G. Cao and A. Luo, Effect of urban thermal characteristics on wetlands based on remote sensing and GIS, Urban Remote
Sensing Event, (2009) Joint, 1–6.
[4] N. Schwarz, U. Schlink, U. Franck and K. Grossmann, Relationship of land surface and air temperatures and its implications for quantifying
urban heat island indicators – An application for the city of Leipzig (Germany), Ecological Indicators 18 (2012) 693–704.
[5] T. Katayama, T. Hayashi, Y. Shiotsuki, H. Kitayama, A. Ishii, M. Nishida, J. Tsutsumi, M. Oguro, Cooling effects of river and sea breeze on
thermal environment in a built up area, Energy and Buildings 15-16 (1990/1991) 973-978.
[6] S. Murakawa, T. Sekine and K. Narita, Study of the effects of river on thermal environment in an urban area, Energy and Buildings 15-16
(1991) 993-1001.
[7] N. Nishimura, T. Nomura, H. Iyota and S. Kimoto, Novel water facilities for creation of comfortable urban micrometeorology, Solar Energy
64 (1998) 197–207.
[8] E.A Hathway and S. Sharples, The interaction of rivers and urban form in mitigating the Urban Heat Island effect: A UK case study, Building
and Environment, 58 (2012) 14-22.
[9] S. K Jusuf, W. N. Hien, N. I. Syafii, Influence of Water Feature on Temperature Condition Hot Humid Climate, Presented in iNTA-SEGA
2009 - Bridging Innovation, Technology and Tradition, Thailand, 2009.
[10] N. E. Theeuwes, A. Solcerova and G.J. Steeneveld, Modeling the influence of open water surfaces on the summertime temperature and
thermal comfort in the city, Journal of Geophysical Research: Atmospheres 118 (2013) 8881-8896.
[11] M. Robitu, C. Isard, D. Groleau and M. Musy, Energy balance study of water ponds and its influence on building energy consumption,
Building Services Engineering Research and Technology 25 (2004) 171-182.
[12] M. Robitu, M. Musy, C. Isard and D. Groleau, Modelling the influence of vegetation and water pond on urban microclimate, Solar Energy
80 (2006) 435-447.
[13] M. Kanda, T. Kawai, K. Narita, A. Hagishima and R. Moriwaki, A comprehensive outdoor scale model experiment for urban climate, in:
Proceedings of the International Conference on Urban Climate. International Association for Urban Climate: Goteborg, Sweden, 2006, pp
270–273.
[14] M. Kanda, Progress in the scale modelling of urban climate: Review, Theoretical and Applied Climatology 84 (2006) 23-33.
[15] M. Kanda, M. Kanega, T. Kawai, R. Moriwaki, and H. Sugawara, Roughness lengths for momentum and heat derived from outdoor urban
scale models, Journal of Applied Meteorology and Climatology 46 (2007) 1067–1079
[16] T. Kawai and M Kanda, M, Urban Energy Obtained from the Comprehensive Outdoor Scale Model Experiment. Part 1: Basic Features of
the Surface Energy Balance, Journal of Applied Meteorology and Climatology 49 (2010) 1341 - 1359 ͒
[17] T. Kawai and M. Kanda, M, Urban Energy Obtained from the Comprehensive Outdoor Scale Model Experiment. Part 2: Comparisons with
Field Data Using an Improved Energy Partition, Journal of Applied Meteorology and Climatology 49 (2010) 1360 - 1376